linux-hardening - Re: [PATCH] mips: boot/compressed: use __NO

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 31 Oct 2022 00:32:08 -0700
From:   Kees Cook <keescook@...omium.org>
To:     John Thomson <git@...nthomson.fastmail.com.au>
Cc:     tsbogend@...ha.franken.de, linux-mips@...r.kernel.org,
        linux-hardening@...r.kernel.org, kernel test robot <lkp@...el.com>
Subject: Re: [PATCH] mips: boot/compressed: use __NO_FORITFY

On Mon, Oct 31, 2022 at 04:36:47AM +1000, John Thomson wrote:
> In the mips CONFIG_SYS_SUPPORTS_ZBOOT kernel, fix the compile error
> when using CONFIG_FORTIFY_SOURCE=y
> 
> LD      vmlinuz
> mipsel-openwrt-linux-musl-ld: arch/mips/boot/compressed/decompress.o: in
> function `decompress_kernel':
> ./include/linux/decompress/mm.h:(.text.decompress_kernel+0x177c):
> undefined reference to `warn_slowpath_fmt'
> 
> kernel test robot helped identify this as related to fortify. The error
> appeared with commit 54d9469bc515 ("fortify: Add run-time WARN for
> cross-field memcpy()")
> Link: https://lore.kernel.org/r/202209161144.x9xSqNQZ-lkp@intel.com/
> 
> Resolve this in the same style as commit cfecea6ead5f ("lib/string:
> Move helper functions out of string.c")
> 
> Reported-by: kernel test robot <lkp@...el.com>
> Fixes: 54d9469bc515 ("fortify: Add run-time WARN for cross-field memcpy()")
> Signed-off-by: John Thomson <git@...nthomson.fastmail.com.au>
> ---
> not sure about the fixes tag:
> compile error only occurs due to the fortify commit, but it looks like
> this change could have been part of the other commit identified in the
> message: lib/string move helper functions?

I think the Fixes tag is reasonable. Strictly speaking, it should likely
be whatever added the memcpy() in decompress.c, but that's mostly
nonsense, since nothing else had tripped yet. :)

The issue is that the memcpy() has a runtime length, so
CONFIG_FORTIFY_SOURCE was inserting the logic for a runtime warning,
which doesn't exist here. This is the correct fix (turning off FORTIFY
in the early boot loader).

Reviewed-by: Kees Cook <keescook@...omium.org>

-- 
Kees Cook